© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 12553
    Prev
    Next
    INDEX
    Explanations

    phrases related to discussions or arguments about specific topics

    oai_token-act-pair · gpt-3.5-turbo

    discussions involving opinions or reviews

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ´
    -1.07
    âĢ
    -0.97
     Whilst
    -0.89
    ¨
    -0.88
    ãĢ
    -0.87
    ________________________________________________________________
    -0.86
    ãĢį
    -0.85
    ————————————————
    -0.84
    """
    -0.83
    âĢIJ
    -0.82
    POSITIVE LOGITS
     lousy
    0.92
     Slate
    0.76
     Wired
    0.76
     twent
    0.75
     sucker
    0.75
     Greenberg
    0.68
     Manhattan
    0.68
     Xer
    0.67
     typew
    0.67
     quiz
    0.67
    Activations Density 1.351%

    No Known Activations